Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 1987 · 255 citations · 0 references
ComputingAutomotive EngineeringComputer ScienceAutonomous VehiclesComputer VisionMachine VisionCurvature AnalysisLane DetectionMotion EstimationRobotic VisionVehicle PerceptionVisual TrackingAutonomous Vehicle GuidanceCurvatureCurvature-based ModelingReal-time Visual Feedback ControlVehicle GuidanceVision-based GuidanceVisual FeedbackVisual Feedback Processing
The method targets well‑structured highways with clear lane markings. The system employs contour correlation, high‑order world models, perspective projection, and Kalman‑filter dynamics on a multi‑microprocessor platform to generate visual feedback control. The approach was shown to guide high‑speed vehicles up to 100 km/h, eliminating steady‑state curve errors and demonstrating real‑time performance in a video‑recorded simulation.
An efficient method for guiding high speed land vehicles along roadways by computer vision has been developed and demonstrated with image sequence processing hardware in a real-time simulation loop <sup>1</sup>. The approach is tailored to a well structured highway environment with good lanemarkings. Contour correlation and high order world models are the basic elements of the method, realised in a special multi-microprocessor (on board) computer system. Perspec-tive projection and dynamical models (Kalman filter) are used in an integrated approach for the design of the visual feedback control system. By determining road curvature explicity from the visual input, previously encountered steady state errors in curves are eliminated. The performance of the system will be demonstrated by a video film. The operation of the image sequence processing system has been tested on a typical Autobahn-scene at velocities up to 100 km/h.